Alcohol use among persons with HIV exacerbates health problems and accelerates HIV disease progression. Antiretroviral therapy (ART) is the single most important treatment for people living with HIV. However, ART adherence is suboptimal among adolescents and young adults living with HIV (hereafter called youth), the age group with the fastest growing rates of HIV infection, and great risk of engaging in risky behaviors such as alcohol use. The proposed study will compare the effectiveness of home-based versus clinic-based Healthy Choices, a brief, 4- session intervention using Motivational Enhancement Therapy (MET) to address alcohol use, medication adherence, and health outcomes in youth living with HIV (YLH) using a repeated measures design. Unlike previous trials, Healthy Choices will be tested in a real world clinical setting and be delivered by community health workers (CHW), who are already members of the HIV care team. The study population will consist of YLH, ages 16 to 24, who are current patients at 5 ATN sites. Site staff will recruit potential participants in clinic or by telephone. Youth will be randomized to receive Healthy Choices, 4-sessions of MET for YLH, either clinic-based or home-based delivered by the same CHW in both conditions. Outcomes are measured at baseline, 4 months (immediately post-intervention), 7 months (3 months post-intervention) and 13 months (9 months post-intervention). Data collection for biological measures will be conducted through medical record extraction, and self-reported measures will occur using a brief Web-based CASI (computer-administered self-interviewing) survey on an iPad at each site. All intervention sessions will be audio-recorded for MITI fidelity coding, and investigators will support local supervisors during the active intervention phase. We will conduct qualitative interviews with CHWs, supervisors and organization leaders at the end of the trial to obtain critical information about barriers and facilitators of implementation. Thus, the proposed trial will allow us to use a Type 1 Effectiveness-implementation hybrid design to pilot a sustainable model of MI implementation in real-world youth care settings towards the goals of 1) examining the effectiveness, cost-effectiveness, and scalability of an efficacious behavioral intervention when delivered by CHWs in real-world adolescent HIV care settings; 2) gathering information about who responds under what contexts; and 3) increasing our understanding of the barriers and facilitators for future implementation.